Atlas

Compare Models

Gemini 4 Neon vs Muse Ember

Gemini 4 Neon leads by 6.8 capability points (≈95% interval +4.0 to +9.6): a clear lead over Muse Ember · P(higher) ≈ 100%

Wins 47 of 49 shared benchmarks (2 lost)

Capability vs. relative cost
Gemini 4 NeonMuse EmberClosed weightsOpen weights

Benchmark wins47 – 2 · 49 shared

Comparing matching setups only (2 benchmarks excluded) · suites counted once
← Gemini 4 NeonMuse Ember →
Data: Artificial Analysis, Vals AI, Surge AI and 3 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Ember minus Gemini 4 Neon.

Terminal-Bench-Science 0.1Science › Multidisciplinary70.017.6−52.4
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge71.325.1−46.2
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents18.41.0−34.0
RiemannBench (November 2026 rescoring)Math84.053.6−30.4
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine47.217.2−30.0
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering75.847.1−28.7
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering63.635.4−28.3
GDP.xlsxProfessional › General knowledge work43.115.4−27.7
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax50.524.2−26.3
Artificial Analysis Intelligence Index v5.0.1Composite indexes61.436.0−25.4
Artificial Analysis Legal IndexProfessional › Legal63.838.5−25.3
Artificial Analysis Engineering IndexScience › Engineering66.841.8−25.1
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents44.5-2.7−23.6
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge54.87.7−23.5
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations52.329.8−22.5
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering44.422.0−22.4
Terminal-Bench 4.0Software › Software engineering61.639.4−22.2
SimpleBenchKnowledge & reasoning › General knowledge94.472.7−21.7
Artificial Analysis Economics IndexComposite indexes61.941.0−21.0
LMArena Code Arena WebDevSoftware › Frontend & design17641608−20.7
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax65.144.6−20.5
Artificial Analysis Coding Agent Index v2.1Software › Software engineering58.639.7−19.0
Artificial Analysis Healthcare & Medical IndexHealth & medicine60.141.3−18.8
CritPt v2 (agentic, avg@5)Science › Physical sciences77.462.0−15.4
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations77.462.1−15.3
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work63.848.7−15.1
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following72.958.8−14.1
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine31.718.3−13.3
APEX-Agents 1.1Professional › General knowledge work88.975.7−13.2
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations79.368.1−11.3
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work33.422.8−10.6
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents19.2-2.0−10.6
Vals Index v2.2Composite indexes70.960.5−10.4
Vibe Code Bench v1.1Software › Frontend & design93.783.7−10.0
AA-Briefcase v1.1 EloProfessional › General knowledge work16991526−9.5
Tax Agent BenchProfessional › Finance, accounting & tax79.871.4−8.4
LMArena Agent Arena Steering BurdenTool & computer use › General agents21.24.8−8.2
Vals Finance IndexProfessional › Finance, accounting & tax68.360.9−7.4
Chartography (November 2026 rescoring)Perception & media › Documents & charts86.679.3−7.3
Legal Research BenchProfessional › Legal56.349.0−7.2
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents-6.06.6+6.3
τ³-Banking (Mercor run, bm25_grep retrieval)Tool & computer use › Customer service & operations59.553.3−6.2
HLAB-Vals-1.1Professional › Legal19.625.0+5.4
Finance Agent v2.1Professional › Finance, accounting & tax68.663.8−4.7
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax76.572.2−4.3
GDP.pdf (November 2026 rescoring)Professional › General knowledge work53.449.2−4.2
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Professional › Legal26.622.7−3.9
Vals Legal IndexProfessional › Legal78.475.7−2.7
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.00.90.0
Data: Artificial Analysis, Vals AI, Surge AI and 3 othersatlas.village.dev
Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…